Abstract
Stroke is a leading cause of mortality and long-term disability worldwide, characterized by substantial molecular and clinical heterogeneity. This study aimed to investigate the metabolic and immune-related mechanisms underlying stroke subtypes using transcriptomic data and advanced computational tools. Gene expression data from two GEO datasets (GSE16561 and GSE58294) were preprocessed, batch-corrected, and integrated. Consensus clustering based on 2,752 metabolism-related genes identified three distinct subtypes with significant differences in metabolic activity. GSVA analysis revealed subtype-specific variations in key pathways, such as the citrate cycle, glycolysis, and glycosaminoglycan biosynthesis, highlighting their metabolic diversity.